Attitudes Towards Artificial Intelligence at Work Scale (AAAW)

Description

The Attitudes Towards Artificial Intelligence at Work Scale (AAAW; Park, Woo, & Kim, 2024) was designed to assess workers’ attitudes towards the application of artificial intelligence (AI) in the workplace. It evaluates six dimensions: perceived human likeness of AI, perceived adaptability of AI, perceived quality of AI, AI use anxiety, job insecurity, and personal utility of AI. The scale was developed using items adapted from the literature on technology acceptance, human-computer interaction, and AI characteristics. Data were collected from three independent samples of working adults, and the scale underwent rigorous testing for factor structure, reliability, and validity. This measure, consisting of 25 items, provides insights into the multifaceted attitudes workers have towards AI integration in their work environment.

Author

• Park, Jiyoung ORCID
• Woo, Sang Eun
• Kim, JeongJin ORCID

Purpose

The scale is intended to assess attitudes towards AI applications in the workplace, specifically focusing on how employees perceive various dimensions of AI, including its adaptability, humanlikeness, quality, and its impact on job security and personal utility. This tool helps understand workers’ psychological responses and behaviors regarding AI integration into their jobs.

Construct

The six dimensions measured by this scale are:

• Perceived Humanlikeness of AI
• Perceived Adaptability of AI
• Perceived Quality of AI
• AI Use Anxiety
• Job Insecurity
• Personal Utility of AI

Instrument Type

Inventory/Questionnaire

Items of the “Attitudes Towards Artificial Intelligence at Work Scale”

• Number of Items: 25
• Rating Scale: 5-point Likert scale (1 = strongly disagree, 5 = strongly agree)
• Format: Electronic

Reliability and Validity

Internal Consistency

• Cronbach’s alpha for the six dimensions exceeded 0.80, indicating strong internal consistency.
• The composite reliability (CR) values ranged from 0.70 to 0.94, all exceeding the cutoff of 0.70, indicating good reliability.

Validity

• Convergent Validity: The average variance extracted (AVE) values for the six dimensions ranged from 0.52 to 0.77, exceeding the threshold of 0.50.
• Discriminant Validity: All factor correlations were smaller than the square root of the AVE, supporting discriminant validity.
• Criterion-related Validity: The scale explained unique variances in recruiting outcomes, supporting criterion-related validity.
• Construct Validity: Factor analyses confirmed the structural validity of the model.

Factor Analysis

• Exploratory Factor Analysis (EFA): A six-factor solution was found to be both interpretable and parsimonious, retaining 43 items across the six dimensions, which were then refined into 25 items.
• Confirmatory Factor Analysis (CFA): The six-factor model showed a good fit (χ2 = 673.60, df = 260, CFI = 0.98, NFI = 0.96, TLI = 0.97, RMSEA = 0.04, SRMR = 0.04). A second-order factor model also provided good fit indices, though it was found that the six-factor model was a better fit.

Test Methodology

The scale underwent extensive validity testing including construct, convergent, criterion, and discriminant validity, along with factor analyses to assess its reliability and fit to the data.

Items of “Attitudes Towards Artificial Intelligence at Work Scale”

The items on this scale explore various worker attitudes towards AI in the workplace, focusing on perceptions of its humanlikeness, adaptability, quality, and its effects on job anxiety and security, as well as its utility in personal and professional contexts.

Reference

Park, J., Woo, S. E., & Kim, J. (2024). Attitudes towards artificial intelligence at work: Scale development and validation. Journal of Occupational and Organizational Psychology, 97(3), 920–951. https://doi.org/10.1111/joop.12502